TrustDiscussion

MeatballWiki | RecentChanges | Random Page | Indices | Categories

BrainstormMode. With WhatIsTrust and DistributedTrust a lot of questions turn up.

If there are so many different definitions and concepts of trust, can we be sure that the word "trust" means a single phenomenon? Maybe its just a word used for different, although related, things? Does "trust in God" really refer to the same thing as "trust a technical system" and "trust a (wo)man".

The difference between trust that grows as a potential and trust that is actually given in actions seems blurred. There is also unjustified trust. There is also trust that shows from actions, but people are not aware of the trust (the risk) they implicitely give.

Trust seems tightly related to the freedom of decisions. If you don't have a choice, the concept of trust makes little sense. Maybe trust means the willingness to take a risk to get an advantage. This is related to cooperation. If you have no choice, one can still "speak of trust" and hope for the best. There is also "trust in your luck".

Trust - cooperation - risk - advantage. Maybe its hard to build a complete picture without building a complete system or model of terms. WhatIsTrust contains the word "risk" and "cooperation" once (literature) and "advantage" never.

Trust and risk may be absolute or relative. Two people A and B both can trust a person to lend him $100, but for A it's a big thing while B is a millionaire and $100 are peanuts. It's not necessarily about money, maybe it's time or information or intimacy, but it depends on ones richness and ones willingness to separate from some of this richness. Maybe trust can't be separated from issues of ownership and richness and protecting forces like institutions and one's own strength.

The problem of trust may point to one of the problems of science: its analytical nature. The big picture is rarely targeted in science. Instead, scientists prefer to specialize and work about the details of details. In the scientific system this is more rewarding, because it's easier to get at publishable news. At a point, papers are counted, their value is not.

My definition of trust was for a paper in a class where I am not allowed to speak my own mind, so that's something to note. However, thinking about "WhatIsTrust" I think the frame of the definition is somehow not right. I think it's better to look at trust as what function it serves to even have trust.

We have trust in order to do things larger than TheIndividual. All structures with formerly independent parts exhibit trust. The cells in your body trust each other not to metastatize to try to overwhelm the rest of the body. To EnforceResponsibility, we have an immune system to act as a PoliceForce. Thus, trust what allows more than one individual to work together in some way.

I think a lot of people look at coercion as the opposite of trust. That is, if I hold a knife to your neck and tell you what to do, I am saying I distrust you so I must coerce you to behave. Some definitions of trust therefore suggest that trust relationships do not have power, but that is not really true (e.g. parent vs. child). Perhaps it is the case that coercion creates 'trust', in the sense that the only way I can predict your actions is by threatening your life, and thus I can trust that you will behave again. Perhaps the confusion is that initially I had distrusted you, but with the introduction of violence I can once again trust you, albeit at extreme cost.

The reason why the knife example is important is that NonViolent trust can only happen in a SafePlace?, which often means there needs to be a PoliceForce to maintain the CommunityBoundary? of the SafePlace?, and the PoliceForce is akin to the knife, except it is indirect to our particular trust relationship. Also, SocialRecourse? such as lawsuits create trust because of the threat of reprisal. So, I think, in short that we have mechanisms to change distrusted relationships into trusted relationships, and these mechanisms have a cost, and thus we prefer more trustful relationships because they have a lower cost. That point is interesting in the context of Meatball since we can design and build these mechanisms, and also analyze mechanisms like PGP in that framework. -- SunirShah

Trust seems to me to be related to the expectation(s) of a result more than it does to a subjective ethical judgement of the result (of an interaction or a transaction between parties). In effect, I trust someone to act in a predictable manner (perhaps even with the expectation that I will consider the result to be 'evil').

A preliminary musing, while preparing to kick this subject around a bit. Given the work that has already gone into WhatIsTrust, it may be quite a challenge to craft a meaningful, incremental contribution. -- HansWobbe